The Sit-Down: Eric Phillips, Delta Air Lines
Pricing and revenue management is obviously something that’s been very unique to the airline industry. It’s interesting to see how it’s moved across industries over the last few years.
In airlines, especially in pricing and revenue management, we are maybe the originators of big data, especially when you consider the scale that Delta flies.
We set the schedule, sell the schedule, then run the schedule.
The whole strategy that we’ve employed, as an airline, is that we want to be the No. 1 airline, the preferred airline, of the business traveler. Thus, it caused us to change the way we thought about what we sold.
We didn’t sell just flights, and we don’t sell just seats. We really sell time. … Because that is what the customer, whether it’s a leisure customer or a business traveler, really values.
If [our schedule] is very, very reliable, it will change how the customer thinks about what flights they purchase when they have other options, and what price they’re willing to pay.
Our job is to find the price point that the customer is willing to pay, not want to pay.
“Willing to pay” is a very different type of conversation that we have with customers, depending on why they’re traveling, where they’re going, when they’re going.
You have to then think about the relationship with the customer beyond then, just that one transaction, just that one trip.
Pricing is always thinking about customer segmentation. The more granular we can get about customer segmentation, and really customer segmentation pertains to the behaviors of the customer.
The way to think about what inventory management is doing is we’re always balancing between yield … and load factor, how full are we.
We could be 100 percent full if all of our fares were $50. Conversely, we could flip it, we could say let’s make every seat $1,000. And we would find a few people who were willing to pay that, but our load factor would be really low.
“What do you look for in job hires?” … The one that we look for in pricing and revenue management is risk-reward trade-off. If you can find an analyst who’s really, really comfortable with risk versus reward, and making that trade-off pretty easily and quickly, those are people who are really, really good at their jobs in pricing and revenue management.
It’s really fascinating. There’s big algorithms we’re running behind the scenes.
The Super Bowl is amazing, but it also gets really complicated for us because it moves. So our demand forecaster will look at a city and be like, “That day in February was awesome. Let’s wait next year for all of this really great demand that’s going to show up, and all these really high price points.”
And the analyst has to be like, “No, no, no, no, you don’t get it. That’s not going to happen this year. We’re going to have to think about it a little bit differently.”
We made a move about six years ago to sell more of our first-class seats, especially domestic.
The front cabin for us domestically is always full. Upgrades are very, very important to people, upgrades are valuable. So while the front cabin was always full domestically, only about 13 percent of the people who sat there had paid for it.
Rather than the fare being six times what a coach fare was … what if we just changed the proposition and said, “Would you be willing to pay us an incremental amount more than the coach fare to confirm yourself into the front cabin?”
The cool thing about the incremental amount is that it would change. If you were buying a discounted coach fare and it was a leisure product, that incremental amount would be pretty big. If you were buying a business product in the coach cabin, that incremental amount shrunk.
We started this about six years ago. Right now our paid load factor in the front cabin is just north of 60 percent for domestic.
Airline seats have always traditionally been thought of as table salt. It’s a commodity. A seat is a seat is a seat.
We want to make it really seamless and simple for the customer to be able to see what their options are and buy it and then make changes to it later if they want.
When Georgia played Notre Dame on Saturday (Sept. 9) and then the Falcons played the Bears on Sunday at Soldier Field, that was a really unique experience for sports fans, to be able to pair up two events like that. We saw an incredible amount of demand for that.
I think that there’s probably room, though, on kind of an everyday basis, more room for partnership between the sports teams and event promoters and airlines.
We use a lot of artificial intelligence already, in terms of the algorithms that we have, but we always know the relationships we’re looking for. It’s not like we just put a bunch of data into it and say, “What can you come up with?”